WhatsApp RAG chatbot with Supabase, Gemini 2.5 Flash, and OpenAI embeddings
Transform WhatsApp into a powerful AI assistant that provides instant answers based on your private document library using advanced RAG technology. By combining Gemini 2.5 Flash with Supabase vector storage and OpenAI embeddings, this workflow enables sophisticated retrieval-augmented generation for real-time mobile support. It offers a complete no-code solution for building a portable knowledge expert accessible anywhere via chat.
Run this with your team's AIWhat This Recipe Does
Provide instant, accurate responses to your customers on the platform they use most. This automation transforms WhatsApp into a sophisticated 24/7 support agent powered by your own business data. Unlike standard chatbots that rely on rigid scripts, this solution uses Retrieval-Augmented Generation (RAG) to pull relevant information directly from your internal documents, manuals, or product catalogs. This ensures that every answer provided is grounded in your specific business context, significantly reducing the risk of incorrect information. By integrating Gemini 1.5 Flash with high-performance vector storage, the system understands complex queries and delivers human-like responses in seconds. This eliminates the bottleneck of manual support, allowing your team to focus on high-value tasks while your customers receive immediate assistance. Whether you are handling product inquiries, troubleshooting common issues, or providing service details, this automation maintains a consistent brand voice and high accuracy. It effectively bridges the gap between your static knowledge base and active customer engagement, turning every WhatsApp message into an opportunity for efficient service.
What your team gets
Forms and dashboards, so it is not a script only one person understands
Runs on your schedule in the cloud, so it does not stop when a laptop closes
Endpoints, so the rest of your stack can trigger the same work
Langchain.vectorStoreSupabase, WhatsAppTrigger, Switch, Langchain.documentDefaultDataLoader, Langchain.embeddingsOpenAi connected for the team, not per person
How It Works
- 1
Open the recipe and connect your accounts
Connect Langchain.vectorStoreSupabase and WhatsAppTrigger once, in your team cloud, and nobody has to do it again on their own machine
- 2
Tell your own agent what is different about your process
Claude, ChatGPT, Cursor, whichever your team already uses. It adapts the recipe to how you actually work
- 3
Run it, then leave it running
It lives in your team cloud, so it keeps going after you close the laptop and every teammate's AI can use it
Who Uses This
- Customer Support Teams use this to automate answers to frequently asked questions based on their latest product manuals and help articles.
- Real Estate Agencies allow potential buyers to inquire about property details and availability directly through WhatsApp using their current listings database.
- Internal HR Departments provide employees with an instant way to ask questions about company policies, benefits, and procedures without manual intervention.
Frequently Asked Questions
What do I need to get started?
You will need a WhatsApp Business API account and your business documents uploaded to a Supabase database to serve as the chatbot's knowledge source.
How does the bot stay accurate?
The system uses OpenAI embeddings to search your specific data, ensuring the AI only answers based on the information you provide rather than general knowledge.
Can I customize the bot's personality?
Yes, you can define specific instructions for the AI to ensure the tone and style of the responses match your brand's voice.
Is my business data secure?
Your data is stored in your private database instance, and the automation only accesses the specific information required to answer the user's question.
Coming from n8n?
This recipe uses nodes like Langchain.vectorStoreSupabase, WhatsAppTrigger, Switch, Langchain.documentDefaultDataLoader and 6 more. On Runwork, you don't need to learn n8n's workflow syntax. Describe what you want to your own AI agent in plain English.
Based on n8n community workflow. View original
Related Recipes
Telegram user registration workflow
This automation streamlines the bridge between AI-driven communication and structured data management. By connecting Telegram interactions with Google Sheets, it transforms unstructured chat messages into organized, actionable records. The workflow acts as an intelligent intermediary that receives data via specialized triggers, processes the information through conditional logic, and ensures every interaction is documented accurately. For businesses, this means eliminating the manual task of copying data from chat apps into spreadsheets. It provides a reliable way to capture leads, log support requests, or collect field data in real-time. By utilizing Runwork to turn this workflow into a dedicated application, your team can manage these data flows through a professional interface without ever touching a line of code or a complex backend. The result is a more responsive operation where information moves instantly from a conversation into your core business systems, improving data integrity and response times.
Create a Slack chatbot with AI for automated responses
This AI-powered automation bridges the gap between conversational intelligence and team collaboration by transforming a standard chat interface into a powerful information distribution hub. By integrating advanced language processing with Slack, this workflow allows your team to interact with an AI assistant that doesn't just answer questions, but actively documents insights and communicates findings across your organization. Instead of losing valuable information in isolated chat windows, this automation ensures that every AI-generated insight is captured as a digital note and shared instantly with the relevant stakeholders. This streamlines internal knowledge sharing, reduces the need for manual status updates, and ensures that critical data derived from AI interactions is immediately actionable. For businesses looking to scale their operations, this tool eliminates the manual overhead of copying and pasting information between platforms, allowing your team to focus on high-level strategy while the automation handles the documentation and notification logistics.
Build a product catalog chatbot with Mistral AI, Google Drive & Supabase RAG
Managing vast amounts of information across Google Drive can lead to significant bottlenecks when teams need quick answers. This automation streamlines the process of transforming static documents into an interactive AI knowledge base. By automatically extracting text from files stored in Google Drive and processing them in manageable batches, the system prepares your proprietary data for use in custom AI chatbots. This eliminates the need for manual data entry or tedious copy-pasting from PDFs and documents. Business leaders can now ensure their AI tools are powered by the most current internal documentation, leading to higher accuracy in automated responses. The workflow handles the heavy lifting of data preparation, allowing your team to focus on high-value analysis rather than document administration. By implementing this solution, you create a scalable bridge between your unstructured files and actionable business intelligence, significantly reducing the time spent on internal information retrieval.
Create a Telegram customer support bot with GPT4-mini and Google Docs knowledge
This automation bridge the gap between instant messaging and formal documentation by transforming Telegram conversations into structured Google Docs records. Instead of manually copying and pasting ideas, meeting notes, or project updates from a chat thread, this workflow captures incoming messages and organizes them directly into your document management system. By automating the transition from a casual chat interface to a professional document format, your team can ensure that critical information is never lost in a busy message history. This tool is particularly valuable for capturing spontaneous brainstorms, field reports, or client requirements in real-time. It streamlines the content creation process, allowing users to focus on communication while the AI handles the administrative task of cataloging and formatting information for future use.
Run this with the AI your team already uses
Your agent adapts it, your team cloud keeps it running, and everyone's AI can find it.
Open this recipe in Runwork